## October 5, 2007

### 2D vs. 3D

Filed under: computer vision/machine vision/AI, mathematics, news — Peter @ 2:36 am

Oh TechCrunch and its confused readers (and writers!)… In a recent post tells about a company that uses “extremely wide angle lenses to capture full spherical images of the urban environment” to create a “3D panorama”. They expect somebody to find “the latitude, longitude, elevation, and other attributes of garbage cans…”. A discussion of whether this is a “true” 3d follows.

As I wrote in my comment, there is an easy way to define 3D: “I can see the object AND I know how far it is”. The object is a 2D picture and the distance (depth) is the 3rd dimension. Without that, it’s not 3D. It does not matter whether the picture is curved. I would even venture to suggest this “rule”:

To capture a 3D image you need a 3D camera.

What is a 3D camera? Well, any camera takes 2D pictures so all you need to add is the 3rd dimension. Time could be that, so a video camera is a 3D camera. Or you could combine several cameras in a row - that row is the 3rd dimension (in fact just two cameras will do). In either case, you can find the distance via stereo vision. Or you could simply add a distance measuring device such as radar, lidar, etc.

The company in question makes thousands of pictures from a moving car, so there is a third dimension. But since it seems that they don’t do any stitching, then maybe not…

### 4 Responses to “2D vs. 3D”

1. motters Says:

Yes this isn’t true 3D, although with enough images it may be possible to give the viewer the impression of 3D. If lidar were used at the same time it would be possible to turn this data into 3D models though.

Another interesting possibility is that if objects of known size/height can be observed this can be used to calibrate the system and obtain structure from motion. The system would only need to be calibrated once to recover structure from an entire sequence of images (see Andrew Davison’s SLAM).

On the first point. My angle here is computer vision. So, even if you can fool people, there is no point in trying fool computers. They don’t pay money.

On the second point. Calibration indeed solves the problem of 3rd dimension. The solution however is only as complete as your collection of objects of known size. Imagine going through a forest…

3. Computer Vision for Dummies » Computer vision in TechCrunch awards Says:

[…] Earthmine – reconstructing cities from street views, “first geospatially accurate and complete street-level 3D data”. Well, conversion of 2D to 3D isn’t going to work. If they collect images continuously by driving through the streets and then patching the images together (that’s unclear), they get a 3rd dimension. Even if this is the case, that only gives them horizontal lengths while heights are lost. Bottom line, even their demo shows only static (panoramic) shots not a true 3D reconstruction. […]

4. Computer Vision for Dummies » The biggest commercial success of computer vision ever Says:

[…] BTW, the Roomba does have vision however rudimentary. It does not detect vertical changes, so it is fair to say that its vision is 1-dimensional. Taking time into account it’s 2-dimensional. Another 1D vision system is radar. […]